The current application proposes to evaluate motivational factors associated with recruitment into and efficacy of brief computer-delivered interventions for heavy drinking college students. College students are at increased risk for alcohol misuse compared to other adults, and development of efficacious intervention approaches is an urgent priority for NIAAA. Over the past several years empirical evidence has demonstrated support for computer-based personalized normative feedback (PNF) interventions which correct normative misperceptions of drinking and thereby reduce drinking behavior. Two specific questions which have remained largely unaddressed include whether the effectiveness of computer-based interventions depends on the location in which they are completed, and what is the optimal incentive structure to balance recruitment with motivational biases in intervention trials. A preliminary comparison of findings across several studies suggests that remote PNF interventions may be less effective than in-lab PNF interventions. However, the evidence is not conclusive because no study provides a direct comparison between delivery methods based on random assignment within the same study. This R21 proposes to recruit two cohorts of 240 heavy drinking college students and randomly assign them to a 2 x 2 x 2 repeated measures design. Participants will be randomly assigned to in-lab PNF, remote PNF, in-lab attention control, or remote attention control. Participants will also be randomly assigned to receive no incentive ($0) or an incentive ($30) for participation. Assessments will include baseline, post- intervention, 3-month, and 6-month follow-ups. College students' motivational orientations, incentives, and typical drinking are expected to moderate differences between delivery methods. We expect that students who receive incentives for participating in intervention studies will be more likely to participate, but less likely to reduce drinking because they will be more likely to attribute their participation to extrinsic motivation. Further, students who receive computer-based PNF in the laboratory for no incentive are expected to exhibit the largest reductions in drinking. In contrast, students administered attention-control feedback remotely for an incentive ($30) is expected to exhibit the smallest reductions in drinking. Perceived value of intervention, retention of intervention content, and attribution for participation are expected to mediate incentive effects and differential efficacy of in-lab versus remote delivery. This research is expected to yield theoretical and practical improvements to feedback-based intervention strategies with potential to reduce drinking and related negative consequences with a stronger theoretical basis and at lower cost than have been previously available.